Since Arne Duncan became Secretary of Education and unleashed the Race to the Top, almost every state has adopted laws to evaluate teachers by the test scores of their students. Most teachers know that this is unfair because the factors that have the greatest influence on students’ test scores are not within the control of teachers. Reformers tell us that teachers are the most important influence within the school on student scores, and that is right. But the teacher contribution to scores is dwarfed by the influence of family and other out of school factors.
It is also obvious to everyone but the U.S. Department of Education that when testing becomes the determinant of teachers’ evaluation, their reputation, and their careers, the results are predictable: narrowing of the curriculum, teaching to the test, gaming the system, and cheating. None of this improves education. Why would any responsible public official want to promote such behavior?
The eminent mathematician John Ewing, who is the president of Math for America, wrote a concise and slashing attack on the misuse of mathematics in value-added methodology. He writes about how teachers in Los Angeles were bullied by journalists who ranked them and then confronted them with their low scores. The journalists warned that value-added should not rely on a single measure, but they themselves relied on a single measure to create their rankings.
Ewing says that the public is being subjected to “mathematical intimidation” by policymakers and education “experts,” and that mathematicians have a duty to speak out and tell them to stop misusing their field for political ends.
Just the fact that it is is labeled junk science seems to give it the whiff of maybe misapplied science as opposed to not science at all. Moving numbers around is not science at all. it makes it sexy sounding when the powers that try to be use data and data driven and scientific but it points out how woefully their understanding of science, the scientific process, statistical analysis ( on the crudest level). It is not anything close to science.
It isn’t anything close to math either ( though the mathematicians can speak for themselves once they pull themselves off the floor for crying or laughing at how these people are trying to use math to pull wool over people’s eyes). The closest this comes to is economics as in voodoo economics.
Should I call it “voodoo science” or does that suggest that it is a sort of science when it is just politics covered with data?
Try Junk Politics. That describes just how empty VAM really is, when we know real politics is already vacuous.
This is an article I wrote for the Philadelphia Daily News which points out the dangers of using high-stakes testing to make decisions about the futures of teachers and children.
http://www.philly.com/philly/opinion/142558715.html
“Reformers tell us that teachers are the most important influence within the school on student scores, and that is right.”
No, it’s not right, it maybe correct but it is wrong to use student scores, especially standardized test scores, to tell us anything the student. It informs us, albeit very little, about the “interaction” of the student with the test-the testing event, whether paper and pencil or computer, at a particular point in time. The logical error here, as shown by Noel Wilson in “Educational Standards and the Problem of Error” found at: http://epaa.asu.edu/ojs/article/view/577 , is an “attachment error”. We attach the score of the interaction to the student. We compound that error then by the logical error of mislabeling. By attaching a “score” with a name, proficient, passing, failing, etc. . . to the student we have “doubled” the error.
Logically, any conclusion that is obtained through error will be fraught with error, i.e., invalid. (Not withstanding the fact that by chance one can come up with a correct conclusion using error filled methods just as the blind and asnomic squirrel will occasionally stumble across an acorn.) So to discuss ‘student scores’ and teacher influence/effectiveness is false because the student score cannot logically be imputed to say anything about the teacher nor the student, only about the “testing event”.
Almost all of us have been so inculcated with the idea that a grade, test score, etc . . . says “something” about the student him/herself and then attach that label to the student, that it is blasphemous to question it. To question the supposed “basics” of education discourse is traitorous to education. But in logical reality, we should all be questioning these practices.
Are you aware of the recent ruling that LAUSD use student achievement in evaluating teachers (although it is unclear how that evaluation can/will be done). It is based on a law on the books in CA. Here is the LATimes article: http://articles.latimes.com/2012/jun/13/local/la-me-teacher-eval-20120613
Also interesting in recent ed. lawsuit news is the ACLU suit in Michigan asserting a child’s right to read.
yes. It is still wrong. invalid, inaccurate and unreliable.
John Ewing is a very thoughtful guy. Former President of the Math. Assoc. of Amer. This organization is the primary one dealign with collegiate education in mathematics.
I agree. I am worried about what this ruling does to the landscape of CA reform. This is a Broad funded lawsuit yes? I am watching closely a school district implement merit pay (*cough*cough followed by eyeroll and then fist shaking) and the assessment piece is a disaster. It is a small high school district, and the subject area issues are huge, not to mention a fundamental lack of concern regarding what learning and assessment actually are, and I keep seeing the students get lost in the conversation. Although honestly the district didn’t apply for the grant because they had any illusions about merit pay working – the just wanted the money. When I saw the Times article my heart sank.
Watch where the Broad money goes and you will find an effort to undermine public schools and those who work in them.
Here is info on FLA’s VAM hooey. https://app1.fldoe.org/rules/doc/6A-5.0411_223.pdf
That’s scary stuff!
Just a thought…use VAM to help teachers. I’ve been using a watered-down version of VAM (simple regression) to find which students of mine who are really thriving each year and also identifying those who declined. It provides great direction for my own professional development. VAM is a a great servant but a horrible master. VAM is not perfect. It sometimes difficult to find a pattern in my list of “thrivers” (or “positive deviants”) and “survivors”, but the overall practice helped increase the % of students who exceeded the standards in my classes by nearly 13 points this year.
I appologize for the grammar errors in my last post…as I was responding my 2-year-old daughter was wanting attention too!
Who Can Democrats Trust About The Chicago Teachers Strike?…
“So much for Democratic harmony,” is the way Herold Meyerson chose to start his op-ed in The Washington Post analyzing the ramifications of the current Chicago teachers strike on the well being of more »…